How to create a custom Encoder in Spark 2.X Datasets?

Question

Spark Datasets move away from Row's to Encoder's for Pojo's/primitives. The Catalyst engine uses an ExpressionEncoder to convert columns in a SQL expression. However there do not appear to be other subclasses of Encoder available to use as a template for our own implementations.

Here is an example of code that is happy in Spark 1.X / DataFrames that does not compile in the new regime:

//mapping each row to RDD tuple
df.map(row => {
    var id: String = if (!has_id) "" else row.getAs[String]("id")
    var label: String = row.getAs[String]("label")
    val channels  : Int = if (!has_channels) 0 else row.getAs[Int]("channels")
    val height  : Int = if (!has_height) 0 else row.getAs[Int]("height")
    val width : Int = if (!has_width) 0 else row.getAs[Int]("width")
    val data : Array[Byte] = row.getAs[Any]("data") match {
      case str: String => str.getBytes
      case arr: Array[Byte@unchecked] => arr
      case _ => {
        log.error("Unsupport value type")
        null
      }
    }
    (id, label, channels, height, width, data)
  }).persist(StorageLevel.DISK_ONLY)
}

We get a compiler error of

Error:(56, 11) Unable to find encoder for type stored in a Dataset.
Primitive types (Int, String, etc) and Product types (case classes) are supported
by importing spark.implicits._  Support for serializing other types will be added in future releases.
  ......

Answer 1

I think importing spark.implicits._ where spark is actually the SparkSession will solve your error.

I don’t know if you have imported the implicit encoders?

import spark.implicits._

Just follow the steps in this link:

http://spark.apache.org/docs/2.0.0-preview/api/scala/index.html#org.apache.spark.sql.Encoder

 

Now, as far as I am aware nothing really changed since 1.6. And even if something is changed, your current code should work just fine with default encoders for product types.

To get some insights why your code worked in 1.x and may not work in 2.0.0 you'll have to check the signatures. In 1.x DataFrame.map is a method which takes function Row => T and transforms RDD[Row] into RDD[T].

In 2.0.0 DataFrame.map takes a function of type Row => T as well, but transforms Dataset[Row] (a.k.a DataFrame) into Dataset[T] hence T requires an Encoder. If you want to get the "old" behavior you should use RDD explicitly:

df.rdd.map(row => ???)

For Dataset[Row] map check this Encoder error while trying to map dataframe row to updated row